Copertina

Early learning of the optimal constant solution in neural networks and humans

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Rubruck, Jirko, Bauer, Jan P., Saxe, Andrew, Summerfield, Christopher
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Flexible task abstractions emerge in linear networks with fast and bounded units
di: Sandbrink, Kai, et al.
Pubblicazione: (2024)

Nonlinear dynamics of localization in neural receptive fields
di: Lufkin, Leon, et al.
Pubblicazione: (2025)

Softmax $\geq$ Linear: Transformers may learn to classify in-context by kernel gradient descent
di: Dragutinović, Sara, et al.
Pubblicazione: (2025)

MIQCQP reformulation of the ReLU neural networks Lipschitz constant estimation problem
di: Sbihi, Mohammed, et al.
Pubblicazione: (2024)

Adaptive multiple optimal learning factors for neural network training
di: Challagundla, Jeshwanth
Pubblicazione: (2024)

A constrained optimization approach to improve robustness of neural networks
di: Zhao, Shudian, et al.
Pubblicazione: (2024)

Get rich quick: exact solutions reveal how unbalanced initializations promote rapid feature learning
di: Kunin, Daniel, et al.
Pubblicazione: (2024)

DPG loss functions for learning parameter-to-solution maps by neural networks
di: Castillo, Pablo Cortés, et al.
Pubblicazione: (2025)

Scalable neural network-based blackbox optimization
di: Koratikere, Pavankumar, et al.
Pubblicazione: (2025)

The dynamic interplay between in-context and in-weight learning in humans and neural networks
di: Russin, Jacob, et al.
Pubblicazione: (2024)

Towards graph neural networks for provably solving convex optimization problems
di: Qian, Chendi, et al.
Pubblicazione: (2025)

Bayesian continual learning and forgetting in neural networks
di: Bonnet, Djohan, et al.
Pubblicazione: (2025)

When Are Bias-Free ReLU Networks Effectively Linear Networks?
di: Zhang, Yedi, et al.
Pubblicazione: (2024)

Understanding Unimodal Bias in Multimodal Deep Linear Networks
di: Zhang, Yedi, et al.
Pubblicazione: (2023)

Saddle-to-Saddle Dynamics Explains A Simplicity Bias Across Neural Network Architectures
di: Zhang, Yedi, et al.
Pubblicazione: (2025)

When Representations Align: Universality in Representation Learning Dynamics
di: van Rossem, Loek, et al.
Pubblicazione: (2024)

Algorithm Development in Neural Networks: Insights from the Streaming Parity Task
di: van Rossem, Loek, et al.
Pubblicazione: (2025)

The sampling complexity of learning invertible residual neural networks
di: Li, Yuanyuan, et al.
Pubblicazione: (2024)

Multi-task neural networks by learned contextual inputs
di: Sandnes, Anders T., et al.
Pubblicazione: (2023)

Consistent machine learning for topology optimization with microstructure-dependent neural network material models
di: Vijayakumaran, Harikrishnan, et al.
Pubblicazione: (2024)

Abrupt and spontaneous strategy switches emerge in simple regularised neural networks
di: Löwe, Anika T., et al.
Pubblicazione: (2023)

Quaternion recurrent neural network with real-time recurrent learning and maximum correntropy criterion
di: Bourigault, Pauline, et al.
Pubblicazione: (2024)

What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation
di: Singh, Aaditya K., et al.
Pubblicazione: (2024)

Computing high-dimensional optimal transport by flow neural networks
di: Xu, Chen, et al.
Pubblicazione: (2023)

Evaluating alignment between humans and neural network representations in image-based learning tasks
di: Demircan, Can, et al.
Pubblicazione: (2023)

On the rates of convergence for learning with convolutional neural networks
di: Yang, Yunfei, et al.
Pubblicazione: (2024)

Emergence and scaling laws in SGD learning of shallow neural networks
di: Ren, Yunwei, et al.
Pubblicazione: (2025)

Thermodynamics-informed graph neural networks for real-time simulation of digital human twins
di: Tesán, Lucas, et al.
Pubblicazione: (2024)

Make Haste Slowly: A Theory of Emergent Structured Mixed Selectivity in Feature Learning ReLU Networks
di: Jarvis, Devon, et al.
Pubblicazione: (2025)

Formulations and scalability of neural network surrogates in nonlinear optimization problems
di: Parker, Robert B., et al.
Pubblicazione: (2024)

Simmering: Sufficient is better than optimal for training neural networks
di: Babayan, Irina, et al.
Pubblicazione: (2024)

An analysis of optimization problems involving ReLU neural networks
di: Plate, Christoph, et al.
Pubblicazione: (2025)

Pseudo-differential-enhanced physics-informed neural networks
di: Gracyk, Andrew
Pubblicazione: (2026)

Feature learning is decoupled from generalization in high capacity neural networks
di: Göring, Niclas Alexander, et al.
Pubblicazione: (2025)

Progressive multi-fidelity learning with neural networks for physical system predictions
di: Conti, Paolo, et al.
Pubblicazione: (2025)

The impact of allocation strategies in subset learning on the expressive power of neural networks
di: Schlisselberg, Ofir, et al.
Pubblicazione: (2025)

Convergence of gradient flow for learning convolutional neural networks
di: Diederen, Jona-Maria, et al.
Pubblicazione: (2026)

Automatic debiasing of neural networks via moment-constrained learning
di: Hines, Christian L., et al.
Pubblicazione: (2024)

Bayes-optimal learning of an extensive-width neural network from quadratically many samples
di: Maillard, Antoine, et al.
Pubblicazione: (2024)

Multi-frequency wavefield solutions for variable velocity models using meta-learning enhanced low-rank physics-informed neural network
di: Cheng, Shijun, et al.
Pubblicazione: (2025)